The best approach for keeping false positives low is to employ a predictive model and behavioural profiling to accurately assess what is 'normal' behaviour for a particular customer. Predictive analytic models can evaluate the complex interrelationships between variables such as cheque amount, transaction frequency, transaction velocity, and other less obvious attributes to much more effectively differentiate legitimate from suspect transactions.
Multiple fraud solutions
A holistic view of fraud risk is essential for identifying account takeover, as this type of fraud often starts with a series of small non-financial changes to an account, which can be missed with 'point solutions' that only monitor a single product or channel. Often as many as 15 or more different tools can be in place at a financial institution based on the point of interaction. Not only does this result in an increase in the risk of false positives, but it does not allow the bank to accurately identify the threat of account takeover before it occurs.
Similarly, the lack of a holistic view makes it difficult to enable, identify, segment and target the two percent of customers who represent the highest risk of fraud. In order to streamline this process, FIs should seek out solutions that allow fraud mitigation to take place, but also incorporate predictive scoring and profiling on a single platform that can provide a more comprehensive view of the customer and all of their account and digital activity. If fraud teams can utilise behavioural profiling and the ability to quantify risk mitigation through investigation, the threat of losing customers due to limited visibility of the relationship can be minimised.
All customer accounts and activities present different risk levels. It is vital for FIs in Asia to understand and tailor their fraud and risk reviews individually. FIs need systems whereby they can develop a profile for each customer to learn what indicates a normal activity for each individual customer, and hence in turn, identify the risk level of transactions based on a number of key metrics. In addition, it is essential to aggregate each customer's data from a vast array of sources such as their personal and household data, information on their assets and also, information on their previous purchases. By doing so, FIs can create a central repository to manage customers' risk effectively as they will have a more comprehensive understanding of their customers.
Better Protecting Banking Customers in Asia
To maximise customers' banking potential, financial institutions have adopted a full spectrum of banking solutions via online and mobile channels. As such, banks and their customers are becoming more vulnerable to fraud. With emerging payment channels, fraud patterns are often not yet well understood. Therefore, it is not surprising that upon launch of a new payment technology or offering, it is not long before fraudsters will be at work to exploit any new vulnerabilities.
Sign up for CIO Asia eNewsletters.